florida passive nitrogen removal study · present worth (pw) and uniform annual cost (uac) were...

Florida Passive Nitrogen Removal Study

Final Report

Prepared for:

Florida Department of Health Division of Environmental Health

Bureau of Onsite Sewage Programs 4042 Bald Cypress Way Bin #A-08

Tallahassee, FL 32399-1713

FDOH Contract CORY6

By:

Daniel P. Smith, Ph.D., PE, DEE Applied Environmental Technology

Richard Otis, Ph.D., PE

Otis Environmental Consultants, LLC

Mark Flint, PE Watermark Engineering Group, Inc.

6/26/2008

AAEETT

Applied Environmental Technology

Florida Passive Nitrogen Removal Study Final Report 6 26 2008

ACKNOWLEDGEMENTS The authors thank Elke Ursin, Eberhard Roeder and Paul Booher, and the Research Review and Advisory Committee of the Florida Department of Health for supporting this project, and Hillsborough County, Florida for providing use of the experimental test site. Report Preparation

Principal Investigator: Daniel P. Smith, Ph.D., PE, DEE Applied Environmental Technology Project Team: Richard Otis, Ph.D., PE Otis Environmental Consultants, LLC Mark Flint, PE Watermark Engineering Group, Inc.

i


TABLE OF CONTENTS

Acknowledgements............................................................................................................... i

List of Tables ........................................................................................................................ v

List of Figures....................................................................................................................... vii

Executive Summary...............................................................................................................viii

Introduction........................................................................................................................... 1

Literature Search................................................................................................................... 2

Passive Nitrogen Removal .......................................................................................... 2

Literature Search Methodology................................................................................... 9

Search Engines and Databases........................................................................... 9

Test Centers........................................................................................................ 9

Personal Contacts............................................................................................... 11

Literature Search Results............................................................................................. 12

Database Structure ............................................................................................. 12

Organization of Reference Electronic Files ....................................................... 12

Review of Passive Nitrogen Removal......................................................................... 14

Unit Operations .................................................................................................. 14

Aerobic (Unsaturated) Filters............................................................................. 14

Anoxic (Saturated) Filters.................................................................................. 22

Heterotrophic Denitrification l .............................................................. 22

Autotrophic Denitrification l ................................................................. 26

Drainfield Modifications.................................................................................... 27

Denitrification in Soil......................................................................................... 28

Approaches to Passive Nitrogen Removal Systems.................................................... 35

Literature Search Conclusions and Recommendations ............................................... 37

ii


TABLE OF CONTENTS (CONTINUED)

Experimental Evaluation ...................................................................................................... 39

Materials and Methods ................................................................................................ 39

Project Site ......................................................................................................... 39

Experimental Treatment Systems ...................................................................... 39

Operation and Monitoring.................................................................................. 42

Analytical Methods ............................................................................................ 44

Results and Discussion................................................................................................ 45

Applied Hydraulic Loading ............................................................................... 45

Septic Tank Effluent .......................................................................................... 46

Applied BOD and Nitrogen Loading ................................................................. 47

Performance of Two Stage Treatment Systems................................................. 48

Performance of Unsaturated Aerobic Filters (Stage 1)...................................... 53

Performance of Anoxic Denitrification Filters (Stage 2)................................... 53

Statistical Tests .................................................................................................. 56

Experimental Conclusions and Recommendations ..................................................... 57

Economic Analysis ............................................................................................................... 60

Economic Analysis Objectives ................................................................................... 60

Design Criteria ............................................................................................................ 60

Life Cycle Cost Analysis............................................................................................. 62

Hardware Costs ........................................................................................................... 64

Primary Treatment and Final Effluent Disposal.......................................................... 64

Media Costs................................................................................................................. 65

Operations and Maintenance Costs ............................................................................. 66

LCCA Results.............................................................................................................. 67

iii


TABLE OF CONTENTS (CONTINUED)

Passive Nitrogen Removal Recommendations..................................................................... 75

Passive Nitrogen Removal System.............................................................................. 75

Recommendations ....................................................................................................... 77

Design ................................................................................................................ 77

Flow Equalization.................................................................................. 78

Stage 1 Filter.......................................................................................... 79

Stage 2 Filter.......................................................................................... 81

Permitting........................................................................................................... 82

Installation.......................................................................................................... 82

Control ............................................................................................................... 83

Maintenance and Monitoring ............................................................................. 83

Replacement of Passive Treatment Media......................................................... 84

References............................................................................................................................. 86

Appendix A Memo to Massachusetts Alternative Septic System Test Center Appendix B Florida Passive Nitrogen Removal Study Citation List Appendix C Passive Nitrogen Technology Appendix D Quality Assurance Project Plan

Appendix E NELAC Certified Laboratory Water Quality Data

iv


LIST OF TABLES

Page Table 1 Search Engines and Databases ............................................................................ 9

Table 2 Search Terms ....................................................................................................... 10

Table 3 On-Site Centers Contacted .................................................................................. 11

Table 4 Individuals Contacted .......................................................................................... 11

Table 5 Organization of Citation Files ............................................................................. 13

Table 6 Summary of Unsaturated Aerobic Media Filters................................................. 16

Table 7 Factors Influencing Performance of Unsaturated Aerobic Filters....................... 19

Table 8 Media Characteristics Influencing Performance of Filters.................................. 20

Table 9 Summary of Saturated Anoxic Media Filters ...................................................... 23

Table 10 Factors Influencing Performance of Saturated Anoxic Filters ............................ 24

Table 11 Total Nitrogen Removals Below Soil Infiltration Zones..................................... 30

Table 12 Estimates of TN Removal Based on Soil Texture............................................... 31

Table 13 Total Nitrogen Removal Found in Various Studies of OWTS............................ 31

Table 14 NRCS Drainage Classes and Descriptions .......................................................... 32

Table 15 Drainage Class and Expected Impacts on Denitrification ................................... 33

Table 16 Procured Filter Media.......................................................................................... 40

Table 17 Configuration of Two Stage Filter Media ........................................................... 43

Table 18 Applied Hydraulic Loading Rate......................................................................... 45

Table 19 Septic Tank Effluent Quality............................................................................... 46

Table 20 Applied BOD and Nitrogen Loading Rates......................................................... 47

Table 21 Nitrogen Species In Filter Influents and Effluents .............................................. 49

Table 22 Two Stage Treatment System Nitrogen Removal Efficiency ............................. 50

Table 23 Field Parameters In Filter Influents and Effluents............................................... 51

Table 24 Stage 1 Nitrogen Removal Efficiency ................................................................. 54

Table 25 Statistical Tests for Effluent Nitrogen Concentrations........................................ 57

v


LIST OF TABLES (CONTINUED)

Page Table 26 Design Factor Options of Alternatives ................................................................ 63

Table 27 Estimated Costs of Treatment Hardware............................................................. 65

Table 28 Estimated Costs of Filter Media .......................................................................... 66

Table 29 Estimated Costs of Operations, Maintenance and Stage 2 Media....................... 66

Table 30 Uniform Annual Cost and Present Worth of Alternatives................................... 68

Table 31 Passive Nitrogen Removal System Cost Breakout.............................................. 70

Table 32 Full Life Cycle Economic Analysis for Passive Nitrogen Removal Systems..... 72

Table 33 Comparison of Total System LCCA for PNRS and RSF.................................... 73

Table 34 Present Worth Cost Comparison of One Pass Aerobic Filters and RSF ............. 74

Table 35 Stage 2 Design Options ....................................................................................... 84

vi


LIST OF FIGURES

Page Figure 1 Two Sludge Denitrification System .................................................................... 3

Figure 2 Simultaneous Denitrification System.................................................................. 4

Figure 3 Experimental Filter System Schematic ............................................................... 41

Figure 4 Hydraulic Loading Rate Applied to Stage 1 Filters ............................................ 45

Figure 5 Rate of STE Total Nitrogen Applied to Stage 1 Filters ...................................... 47

Figure 6 Total Nitrogen in Influent STE and Effluent of Two Stage Filter Systems........ 50

Figure 7 Two-Stage System Effluent Total Nitrogen versus TN Loading........................ 51

Figure 8 Dissolved Oxygen in Effluent of Unsaturated Filters (Stage 1).......................... 52

Figure 9 Dissolved Oxygen in Effluent of Denitrification Filters (Stage 2) ..................... 52

Figure 10 Effluent Ammonia from Unsaturated (Stage 1) Filters....................................... 54

Figure 11 Stage 1 Effluent NH4+-N versus TN Loading ..................................................... 55

Figure 12 Total Inorganic Nitrogen Removal Efficiencies of Two Stage Systems ............ 55

Figure 13 NOx Concentrations in Stage 2 Filter Effluents .................................................. 56

Figure 14 Passive Nitrogen Removal System Schematic.................................................... 61

Figure 15 Basic Design Elements of Primary Treatment and Stage 1 Filter....................... 62

Figure 16 Basic Design Elements of Stage 2 Filter............................................................. 62

Figure 17 Uniform Annual Cost of Alternative Systems .................................................... 69

Figure 18 Present Worth of Alternative Systems ................................................................ 69

Figure 19 Uniform Annual Cost per Volume Treated......................................................... 71

Figure 20 Unit Nitrogen Removal Costs of Passive Nitrogen Systems and RSF ............... 73

Figure 21 Conceptual PNRS Component Placed within Conventional Onsite System ..... 75

vii


Executive Summary Approximately 2.5 million onsite wastewater treatment systems (OWTS) are currently permitted in the State of Florida. Population growth, exurban development trends, and the high cost and sustainability of centralized infrastructure make it likely that distributed infrastructure will continue to be used for the management of a large portion of domestic sanitary water generated in Florida. The vast majority of onsite systems include a septic tank for primary treatment, followed by dispersal into the environment using soil adsorption systems. Nitrogen removal in these typical systems is limited. Nitrogen loading from onsite systems is a potential concern in Florida, depending on the sensitivity of the water environments, the number and density of onsite installations, their proximity to receiving waters, and processes in subsurface soil media. This Florida Passive Nitrogen Removal Study (PNRS) was undertaken to investigate alternative methods to remove nitrogen in onsite systems. A primary consideration was to evaluate systems that were “passive” in nature, with limited reliance on pumping and forced aeration. A guiding principal for the PNRS was the specific definition of a “passive” nitrogen removal system as one that contains only a single liquid pump, no mechanical aerators, and that uses reactive media. The PNRS was specifically intended to perform a literature review of passive nitrogen removal technologies, perform an experimental evaluation of passive systems and candidate media, perform an economic analysis of such systems, and make recommendations regarding deployment of passive nitrogen systems. Literature Review and Database A literature review was conducted to evaluate technologies that can potentially be used in passive nitrogen removal systems. The literature review included searches in scientific and engineering databases, peer reviewed literature, conference and journal proceedings, unpublished reports, vendor-supplied information, World Wide Web searches, and personal contacts with experts in the field. A searchable database of 227 citations was compiled and provided as a project deliverable. The literature review and analysis of “passive” system constraints were used to formulate a two-stage filter strategy for removing total nitrogen from septic tank effluent. Evaluation of key media characteristics resulted in recommendations of specific media to use in the Stage 1 unsaturated aerobic nitrification filter, and in the saturated, anoxic Stage 2 denitrification filter. The literature review included recommendations regarding key design factors of hydraulic loading rate, dosing regime and media depth of the unsaturated Stage 1 filter and filter sizing and residence time in Stage 2. Experimental Evaluation An experimental on-site wastewater treatment system was operated for sixty days to evaluate enhanced nitrogen removal using two-stage passive nitrogen removal systems. Experiments were performed using actual septic tank effluent at a field site in Hillsborough County, Florida. Two of the three two-stage filter systems achieved over 97% total nitrogen removal and 98% total inorganic nitrogen removal, with average effluent ammonia nitrogen and nitrate+nitrate nitrogen concentrations of less than 0.7 mg/L and 0.5 mg/L, respectively. High

viii


ix

nitrogen removal performance was achieved using clinoptilolite and expanded clay media in the unsaturated Stage 1 filter, and elemental sulfur in the anoxic denitrification filter (Stage 2). The experimental evaluation, though of limited duration, verified the potential of the two-stage filter system for total nitrogen removal using passive technology. Economic Analysis A detailed economic analysis was conducted using Life Cycle Cost Analysis (LCCA) to provide equitable evaluation of the cost of alternative passive nitrogen removal systems over their entire life. LCCA included costs for equipment, materials, and installation, energy, scheduled maintenance, and monitoring, media replacement and residuals management. Present Worth (PW) and Uniform Annual Cost (UAC) were developed for twelve alternative configurations of two-stage passive nitrogen removal systems. LCCA results are presented for both total system cost including passive nitrogen removal, primary treatment (i.e. septic tank) and conventional drainfield, and for the passive nitrogen component only. A cost comparison is also provided for a Recirculating Sand Filter, which is a widely applied onsite technology. Recommendations for System Deployment Recommendations are presented for deployment of a two-stage passive nitrogen removal system for single family homes which discharge septic tank effluent (STE) with characteristics typical of single family residences in the U.S. The passive nitrogen component is placed following primary treatment and before the drainfield in a conventional onsite system. Specific recommendations are presented for system design, including flow equalization and storage volume, pumping arrangement, aerobic Stage 1 filter dosing system, media, filter sizing, and underdrain, Stage 2 anoxic filter media and sizing, and hydraulic profile development. Recommendations for permitting include innovative status application including NSF testing, and possible evaluation of drainfield size reduction credits. Installation, control and monitoring recommendations are made which share commonality with typical onsite installations; a twice per year maintenance visit and one per year monitoring frequency are recommended. The recommendations for replacement of denitrification media (Stage 2) are dependent on the need for longer term performance verification of sulfur-based denitrification filters. In addition, it is recommended to investigate the reuse of spent denitrification media within the treatment process or for beneficial agricultural land application. Recommendations for Future Research Additional studies were recommended to address key issues that have direct implications to two-stage filter process performance, design, feasibility, longevity, and economics. It is recommended to extend operation of the systems to provide longer term operating data, to operated the filter systems at higher loading rates, to employ recycle on Stage 1 filters for pre-denitrification, to more fully examine performance and design issues with the denitrification filters, and to examine treatment parameters other than nitrogen. Full scale testing at a single family residence is recommended for a period of at least two years.


Introduction As population growth continues in Florida, so do the potential impacts of on-site wastewater treatment systems to surface and groundwater quality. Nitrogen loading from wastewater treatment systems may be a concern where numerous on-site wastewater treatment and disposal systems (OWTS) are located within sensitive environments. Conventional septic tank and soil adsorption systems rely on biological reactions in porous media (setback layer or unsaturated natural soil) to attenuate nitrogen loadings to ground or surface water. Groundwater nitrate concentrations have been shown to exceed drinking water standards by factors of three or greater at distances on the order of several meters from soil adsorption systems (Postma et al.,1992). In a study at Big Pine Key, Florida, the dissolved inorganic nitrogen (DIN) levels in groundwater contiguous to on-site drainfields were greater than DIN levels at a control location (Lapointe et al., 1990). Groundwater NH3-N levels at Big Pine Key reached 2.75 millimoles per liter (38.5 mg/L), indicating a high fractional breakthrough of ammonia through the on-site treatment system. In another study, conducted on a sandy Florida aquifer system, groundwater levels of both Total Nitrogen and ammonia were elevated above background levels at a distance of 50 meters from a conventional soil adsorption drainfield (Corbett et al., 2002). Available setback distances in Florida locations may often be quite limited, which increases the significance of achieving high nitrogen removal percentages within septic tanks, media filters and other in-tank treatment processes, as well as with in soil treatment units (Siegrist, 2006). A summary review of a wide variety of on-site treatment approaches showed that systems with some degree of “passive” characteristics exhibited Total Nitrogen removal efficiencies of 40 to 75% and produced effluent TN of 10 to 20 mg/L (Anderson and Otis, 2000). FDOH has an interest in exploring the feasibility and practicality of using relatively passive on-site treatment systems to accomplish even higher nitrogen reductions in a cost effective manner. The mission of the Bureau of Onsite Sewage Programs of the Florida Department of Health (FDOH) is “Protecting the public health and environment through a comprehensive onsite sewage program”. FDOH established the Florida Passive Nitrogen Removal Study to identify passive treatment systems that can achieve greater nitrogen reductions than exhibited by conventional septic tank/drainfield configurations. The FDOH is specifically interested in approaches that employ filter media, or reactive filter media, and systems that which eliminate the need for aeration pumps and minimize the need for liquid pumping. The first step of the Florida Passive Nitrogen Removal Study was to identify treatment configurations, reactive and non-reactive media, performance capabilities of new and demonstrated technologies, and factors influencing performance and longevity. The following section describes the results of the literature review and the genesis of the recommended two-stage system for passive nitrogen removal. The experimental evaluation section describes the results of experiments that were performed to verify total nitrogen removal from actual septic tank effluent using passive, two-stage nitrogen removal technology. The economic analysis section presents a detailed life cycle cost analysis of a passive two-stage nitrogen removal system. Finally, the recommendations section provides specific guidance for deployment of passive two-stage nitrogen removal technology for a single family residence.

1


Literature Review Passive Nitrogen Removal The goal of passive nitrogen removal is to provide on-site systems with relatively simple operation and low life cycle costs. Passive nitrogen removal approaches must be cognizant of the speciation of nitrogen (inorganic vs. organic, particulate vs. soluble, oxidized and reduced), the biochemical reaction sequence needed for complete nitrogen removal, and the use of Total Nitrogen as the generally accepted metric of system performance: Total Nitrogen (TN) = Organic N + Ammonia N + Nitrate N + Nitrite N In septic tank effluent (STE), nitrogen is present in organic and ammonia forms, with virtually no oxidized N. Other nitrogen relationships and delineations are listed below. Total Kjeldahl Nitrogen (TKN) = Organic N + Ammonia N Organic Nitrogen = Filtrable Organic N + Non-filtrable Organic N Total Inorganic Nitrogen (TIN) = Ammonia N + Nitrate N + Nitrite N Total Oxidized Nitrogen (TON) = Nitrate N + Nitrite N TN = TKN + TON Conventional unmixed septic tanks provide sedimentation and removal of suspended solids and particulate nitrogen. STE contains ammonia, filtrable (dissolved) organic N, and non-filtrable (suspended) organic N that has not been removed within the septic tank by sedimentation. The use of strainers to treat effluent from septic tanks (also termed STE “filters”) can enhance removal of non-filtrable organic N. Non-filtrable organic N in STE would be removed in media filters by the standard physical filtration mechanisms of straining, impaction and sedimentation within the filter bed. Of great importance to the configuration of passive nitrogen removal systems are biochemical nitrogen transformations. The significant biochemical transformations are listed below in the sequence in which they must generally occur. Hydrolysis converts particulate organic N to soluble organic N, which in turn releases ammonia through ammonification. Both processes can occur in the presence or absence of oxygen. Hydrolysis

Non-filtrable Organic N → Filtrable Organic N

2


Ammonification

Filtrable Organic N Ammonia N →

Removal of total nitrogen in on-site systems requires both nitrification (an aerobic process) and denitrification (an anoxic process). Nitrification must occur first and must be followed by denitrification. Passive denitrification filters cannot treat septic tank effluent without pre-treatment with some type of aerobic treatment. Therefore, if septic tank effluent is considered as the starting point for examining nitrogen reduction strategies, a systems view of nitrification and denitrification may be most beneficial. Nitrification (requires O2)

Ammonia N → Nitrite N Nitrate N → Denitrification (requires electron donor)

Nitrate N Nitrite N → Di-nitrogen (N2) →

Nitrification requires oxygen, while denitrification requires an electron donor. Oxygen for nitrification can be supplied to liquid in septic tanks, pumping tanks, or other treatment tanks using aeration pumps, or by air ingress (assisted or unassisted) into systems containing unsaturated media, such as packed trickling filters, recirculating sand filters, peat filters, textile filters, and the unsaturated zones of drainfields. Here, the unsaturated media are attachment surfaces for nitrifiers and other microorganisms.

To remove nitrogen, both centralized and decentralized wastewater treatment plants must create the conditions necessary to sustain the biochemical reactions required for nitrogen removal. Several different process trains are used in conventional suspended growth wastewater treatment plants, including “two sludge” systems with separate aerobic and denitrifying microbial populations (Figure 1), and “simultaneous” systems (Figures 2) that accomplish both nitrification and denitrification. “Sludge” in this case refers to the active biomass in the process, which provides the treatment. In the simultaneous process the biomass is a mixture of autotrophs

Figure 1. Two Sludge Denitrification System

3


Figure 2. Simultaneous Denitrification System

(nitrifiers) and facultative heterotrophs (organic degraders & denitrifiers) while in the two sludge system, the two groups of microorganisms are separated in different reactors.

The two sludge system can achieve nearly complete nitrogen removal because both the nitrification step and denitrification step can be optimized for removal of organic nitrogen and ammonia (nitrification step) and nitrate and nitrite (denitrification step). However, since the nitrification step removes nearly all the organic carbon, a separate source of organic carbon is required for removal of nitrate and nitrite (Figure 1). Without adequate carbon source, even though removal of ammonia and organic nitrogen may be highly complete, Total Nitrogen removal will be limited. Though the two sludge process has the advantage that it can achieve more complete nitrogen removal, it is very dependent on an external organic carbon source (Bitton, 1994; Degen, et al., 1991; Oakley, 2005). In the simultaneous system, denitrification is achieved by cycling between oxic and anoxic conditions in a single reactor such that nitrification and denitrification is accomplished “simultaneously” (Figure 2). This process occurs in the filter media when wastewater containing ammonium and biodegradable carbon is applied to aerobic soil. In response to the application, facultative heterotrophs quickly degrade the organic carbon and deplete the oxygen in doing so. The ammonium cannot be nitrified under anoxic conditions, so being a positively charged ion; it may be retained within the filter media depending on the cation exchange capacity. This simultaneous process has the advantages of having a continuous supply of organic carbon from the wastewater for the denitrification step, lower oxygen requirements, and it recycles the alkalinity needed for nitrification. However, the amount of denitrification can be limited depending on the frequency and duration of the oxic/anoxic fluctuations within the filter with respect to the reaction rates. In a field study in soil which investigated OWTS design and operation that would maximize denitrification, Degen, et al. (1991) found that this simultaneous process performed best because carbon is the limiting factor for denitrification in soil. A third process model that has been recognized only recently is an anaerobic, autotrophic bacterial process called Anammox. This process is possible when both nitrate and ammonium occur together under anoxic or anaerobic conditions (Van de Graaf et al., 1995; 1996; 1997). In

4


this process, the autotrophs reduce the nitrate to nitrogen gas while utilizing the oxygen from the nitrate to oxidize the ammonium to nitrate. Because the bacteria are autotrophs, no organic carbon is required to sustain this process. Anoxic or anaerobic conditions are necessary because if not, the heterotrophs would oxidize the ammonium removing the energy source from the autotrophs.

Regardless of the types of nitrogen transforming biochemical reactions within the treatment system and their spatial locations, total nitrogen in the effluent will consist of ammonia, nitrate and nitrite, and organic nitrogen. The ammonia nitrogen levels in the effluent from the unit operations preceding the denitrification filter must be consistently at or below target levels for final effluent ammonia nitrogen, since ammonia may behave conservatively as wastewater passes through the anoxic denitrification filter. For passive denitrification filters, solid phase electron donors are employed that provide attachment surfaces for denitrifying microorganisms and electron donor supply through a process of continuous dissolution over extended time periods. While numerous potential solid phase electron donors exist, the most commonly applied have been lignocellulosic materials such as wood chips and sawdust that support heterotrophic denitrification and elemental sulfur (autotrophic denitrification). The total oxidized nitrogen levels in the effluent from the denitrification filter must be consistently at or below target levels for final effluent oxidized nitrogen, which can be established either independently or be apportionment of the target effluent Total Nitrogen among the nitrogen species. The meaning of the term “passive” for nitrogen removal in on-site wastewater treatment systems can then be addressed within the context of overall STE composition, the forms and speciation of nitrogen, and the mechanisms of nitrogen removal. For the Florida Passive Nitrogen Removal Study, a program specific definition for the term “passive” was provided by FDOH:

Passive A type of onsite sewage treatment and disposal system that excludes the use of aerator pumps and includes no more than one effluent dosing pump in mechanical and moving parts and uses a reactive media to assist in nitrogen removal

The definition of a “passive” system placed significant restrictions on the types of onsite wastewater treatment systems than can be considered. The definition precludes the use of aeration pumps within any system component: septic tank, dosing tank or other treatment chambers. Oxygen for BOD removal and nitrification must therefore be supplied by unassisted aeration to an unsaturated media filter that operates as a four phase system: solid media, water, gas phase, and attached biofilm. Wastewater is supplied at the top of the media and flows downward by trickle flow or percolation. This very common approach to onsite wastewater systems is applied in sand filters and in other media filters, providing ammonification and nitrification. Single pass unsaturated media filters can provide some degree of denitrification using wastewater organics. Recirculation of filter effluent to a septic tank chamber or dosing tank can substantially enhance denitrification and produce Total Nitrogen removals of 60% or greater. To

5


achieve higher Total Nitrogen removal percentages and lower effluent TN concentrations, unsaturated filter effluent can be directed to a denitrification filter. Denitrification filters are possibly the only feasible approach to enhancing TN removal in passive onsite systems beyond that achievable by unsaturated filters. Denitrification filters are saturated with water and are three phase systems: solid media, liquid, and biofilm (possible bubble formation from denitrification is considered relatively insignificant). The solid phase contains a reactive solid media that supplies attachment surface and electron donor for denitrifying organisms. The solid phase electron donors that have been most commonly studied are elemental sulfur and cellulosic materials (sawdust and wood chips). Another stipulation of the “passive” definition is that only a single effluent dosing pump be used. The dosing pump must provide adequate head to convey wastewater from the septic tank effluent elevation, through filter media, and presumably to a soil treatment unit. Wherever the single pump is positioned within the treatment train, the movement of wastewater before and after the pump must be by gravity. In addition to hydraulic conveyance, the pump can provide very important treatment features including the ability to pressure dose, the ability used timed dosing, and the ability to spread wastewater uniformly over the entire area of the filter surface. These features have been exploited numerous unsaturated systems such as intermittent sand filters, and are important for efficient treatment. An additional feature afforded by a pump is the ability to recirculate a portion of filter effluent, using various non-powered splitter devices which do not require power or manual operation. Recirculation of the effluent of an aerobic filter effluent (recirculating sand filter for example) increases denitrification using wastewater organics as carbon source, and can substantially increase TN removal efficiency and decrease effluent TN. An additional treatment consideration is alkalinity and the need to maintain appropriate pH conditions for biochemical reactions. Nitrification consumes 7.14 grams of alkalinity as CaCO3 per gram ammonia N nitrified, and nitrifying microorganisms are inhibited as pH decreases below neutral. For an STE containing 45 mg/L TN, required alkalinity is 321 mg/L. The alkalinity of the starting water supply, as augmented by the increase in alkalinity through domestic water use (perhaps 60 to 120 mg/L), must be sufficient to prevent pH decrease and inhibition of nitrification. Nitrogen removal performance of a total nitrogen removal system could be affected by alkalinity of STE and the effects of pH conditions on biochemical reaction rates. If the pH drops in an aerobic filter due to nitrification, then nitrification might not proceed to completion, leaving a high residual ammonia concentration. Ammonia in the effluent of the first stage aerobic filter could largely pass through s second stage anoxic filter, thereby lowering the overall TN removal efficiency. A benefit of recycle around the aerobic filter is that the partial pre-denitrification would be accompanied by the additional benefit of restoration of alkalinity. Alkalinity restoration may become more important in the future as water conservation trends exacerbate the potential of alkalinity to limit nitrification in non-recycle aerobic systems. The potential advantages of recycle in aerobic systems are increased as TN levels increase in STE. The first stage filter must achieve a high degree of BOD and ammonia removal because these components may not be degraded in the second stage anoxic filter environment. Additionally, a

6


high quality first stage effluent will limit the amount of solids and BOD added to the second stage filter. Lower loadings to the anoxic filter should reduce the possibility of channeling and enable better long term performance and lower maintenance needs. Saturated anoxic filters for passive denitrification have far less studied than unsaturated filters. Anoxic filters are usually fully submerged to preclude ingress of oxygen from air. Oxygen in the incoming flow is probably utilized preferentially near the entrance, enabling anoxic conditions to prevail downstream. Denitrifying microorganisms reduce oxidized inorganic nitrogen, predominantly nitrate, to nitrogen gas. The denitrifying microorganisms grow as biofilms on the reactive media, dissolving the reactive media and using it for nitrate reduction. Nitrate is reduced to nitrogen gas, which leaves the reactor dissolved in the liquid effluent or as small bubbles. The principals of porous media biofilm reactors have been well established. Factors that affect performance include the size, specific surface area, tortuosity and porosity of media, average liquid residence time, superficial flow velocity, linear velocity, uniformity of flow (i.e. channeling), mass transfer and biofilm kinetics. A special feature of the passive anoxic filters is the reactive dissolution of the media. The media must supply enough electron donor for denitrification or nitrate removal may decline. On the other hand, if media dissolution is too rapid, media longevity will be reduced and the reactor effluent will contain excess dissolution product (such as BOD for cellulose based media). A solid phase alkalinity supply, such as limestone or crushed oyster shell, may be required to maintain pH. Over long term continuous operation, flow channeling can result in short circuiting, decreased contact time of with biofilms, and decline in performance. A “passive” treatment system for nitrogen removal must be seen as an integrated sequence of unit operations/processes that can achieve the treatment goal. If it is assumed that the starting point is septic tank effluent (STE), then the total treatment system must meet the target treatment goal. The treatment goal could be expressed as the Total Nitrogen (TN) concentration “leaving the treatment system,” or “entering the environment.” Suppose the goal is to achieve a TN of 2.5 mg/L or less before directing the effluent to a soil absorption field. Assuming nitrite levels are negligible, the effluent TN of 2.5 must be apportioned between 1. organic N, 2. ammonia-N, and 3. nitrate-N: Organic-N + NH3-N + NO3-N ≤ 3.0 The biochemical sequence requires ammonification and nitrification before denitrification. For a process with a final treatment step will be an anoxic denitrification filter with reactive media, then attention must be focused on the organic N and ammonia N concentrations in the influent to the denitrification filter. Ammonia levels could increase across the denitrification filter due to ammonification of influent organic N; ammonia levels could decrease across the denitrification filter by nitrification near the inlet using residual dissolved oxygen in the actual denitrification filter influent.

7


An approach to formulation of process objectives is to estimate the effluent nitrate N achievable in anoxic filters, and allocate the remainder of the target effluent TN to the sum of organic N and ammonia N. For a target effluent TN of 3.0 mg/L: TKNallowable 3.0 - NO3-N ≤

An achievable effluent NO3-N of 1.0 mg/L would mandate a TKN of not greater than 2.0 mg/L. The approach would be conservative from the perspective of ammonification in the anoxic filter, which would not change TKN, although some factors such as autolysis could increase TKN. This discussion points to the important need to reduce TKN in the treatment that occurs before the anoxic denitrification filter. Producing TKN less than TKNallowable should be the first priority of the “first stage” of treatment. For “first stage” systems that accomplish denitrification along with nitrification and ammonification in the same process tank or through recirculation, the critical question is: is the effluent TKN less than TKNallowable. From a knowledge of the functioning of aerobic filter systems treating STE, it is hypothesized that optimization of the aerobic treatment process is the most important factor affecting overall nitrogen reduction. This is speculative, because there is limited experience in the coupled operation of aerobic and coupled anoxic filters in passive configurations. If the aerobic process must be optimized, then the single pump that is allowed should be used to supply STE to the aerobic biofilter. The benefits of more frequent doses of lower volume and more uniform flow distribution will accrue to the aerobic filter, and provide a high quality influent to the anoxic biofilter. Using the pump to supply the aerobic biofilter will enable recirculation, which will lessen the nitrate loading to the denitrification biofilter and reduce alkalinity requirements. For low relief Florida environments, the aerobic filter would be placed above grade to enable gravity flow to and through the anoxic filter and then to a soil treatment unit. The following points summarize the needs that must be satisfied by the passive nitrogen removal technology, and factors that influence the overall approach and configuration:

• the biochemical requirement for initial aerobic reactions (ammonification and nitrification), followed by anoxic denitrification, likely in two separate filters;

• a first stage unsaturated media filter allowing air ingress without aeration pumps; • first stage filter to achieve target effluent ammonia and organic nitrogen level; • second stage saturated denitrification filter with reactive solid phase electron donor and

possible alkalinity source; • second stage design to achieve desired effluent oxidized nitrogen level; • provide adequate head for passive media filtration, enabled by only one effluent dosing

pump; • preferred alternative considered dosing pump to first stage unsaturated (aerobic) filter

that enables timed pressure dosing and uniform effluent distribution; • possible recirculation around first stage (unsaturated) filter; • management of any residual materials resulting from filter media replacement, including

cleaning and reapplication, land application, soil conditioner, and construction.

8


Literature Search Methodology Databases and Search Engines CSA Illumina (http://www.csa.com/) and Science Direct (http://www.science-direct.com/) search engines were used to access multiple data bases, as shown in Table 2. The American Society of Agricultural and Biological Engineers (ASABE) Technical Library was queried, as ASABE has been sponsoring an on-site wastewater treatment symposium every three years. Search terms listed in Table 2 were combined using and operator logic in numerous configurations. In addition, Google (http://www.google.com/) and Google Scholar (http://scholar.google.com/) searches were conducted on the World Wide Web, using the same search terms listed in Table 3. Test Centers The on-site centers listed in Table 4 were contacted regarding information on passive nitrogen removal technologies, experience, and theoretical and practical developments. Site visits were made on May 21, 2007 and October 19, 2007 to the Massachusetts Alternative Septic System Test Center on Cape Cod, MA. During these visits, it was determined that many nitrogen removal technologies were being evaluated at the test center that were subject to non-disclosure by center staff. As a result of the first visit, a memo was prepared and addressed to the test center requesting voluntary information disclosure from technology developers using the test center for evaluation of nitrogen removal technologies. A copy of the memo is included in Appendix A. Table 1 Search Engines and Databases CSA Illumina Biotechnology and Bioengineering Abstracts Environmental Sciences and Pollution Management Environmental Engineering Abstracts Pollution Abstracts Science Direct (over 2000 peer reviewed journals) Applied Science and Technology Civil Engineering Abstracts American Society of Agricultural and Biological Engineers (ASABE)

Technical Library

9


Table 2 Search Terms denitrification wastewater on site nitrogen nitrate ammonia nitrification passive septic carbon wood sawdust sulfur organic media filter filtration solid peat filter recirculating filter sand filter coir filter zeolite filter soil denitrification

10


Table 3 On-Site Centers Contacted Massachusetts Alternative Septic System Test Center Rhode Island On Site Wastewater Resource Center Deschutes County Environmental Health Division, Oregon Department of

Environmental Quality (La Pine National Demonstration Project)

National Environmental Services Center Baylor Wastewater Research Program Personal Contacts Personal contacts were made with individuals who are involved with developing, testing, and evaluating technologies for nitrogen removal in on-site wastewater treatment systems. The individuals contacted are listed in Table 5. Valuable insights were gained through discussions and information transfer, and technical reports and information was obtained that was not otherwise available. Table 4 Individuals Contacted Dr. Bruce Lesikar Texas A & M University Dr. Robert Siegrist Colorado School of Mines George Loomis University of Rhode Island George Huefelder Director, Massachusetts Alternative Septic System Test Center Damann Anderson Hazan and Sawyer, Tampa Barbara Rich Environmental Health Division, Dechuttes Co, Oregon Pio Lombardo Lombardo and Associates Dr. Sukalyan Sengupta University of Massachusetts-Dartmouth Paul Hagerty Hagerty Environmental Wesley Brighton Wastewater Alternatives Dr. Martin Wanielista University of Central Florida

11


Literature Search Results Database Structure A database was constructed using EndNote software provided by Thomson Research Soft (http://www.endnote.com/). EndNote is a contemporary and fully supported standard software tool for publishing and managing bibliographies on the Windows and Macintosh® desktops. Endnote allows internal searches using keywords, and Endnote files can be exported for use in other software. The Passive Nitrogen Removal database contains 227 references, which are listed in Appendix B. The Endnote entries include keywords and abstracts for most citations, and URL addresses are provided for numerous citations. The attached CD includes numerous PDFs for cited articles, and PDFs and Word files containing descriptive and performance data for numerous citations. Organization of Reference Electronic Files References were classified according to the nested tree file framework shown in Figure 1. The files in the attached CD are also organized according to the Figure 1 framework. The numbers in the parenthesis of Table 2 are the numbers of citations or supporting documents in each in each folder. The overall organization includes general nitrogen removal in on-site systems, nitrification processes, denitrification processes, and drainfield modifications. Denitrification processes are classified into heterotrophic and autotrophic processes. Heterotrophic processes are subdivided into citations for general cellulosic, cellulosic sources and other carbon sources. The cellulosic folder includes several separate folders for processes of for studies for which several citations of supporting files are available. The autotrophic citations are dominated by sulfur based systems, testifying to the extensive research in this area. As an example, an internal Endnote search using the single search term sulfur extracted 43 entries in the Florida Passive Nitrogen Removal Study Citation List. The search terms organic and carbon each extracted a similar number of citations. The search terms sand filter, peat, and wetland extracted 29, 14 and 12 citations, respectively. The assembled Citation List includes nitrification processes, including recirculation systems. A system using a recirculation pump, such as a recirculating sand filter, would not be “passive” in the sense that a one-pass flow through media filter would be “passive.” In fact, some state regulatory agencies who are considering the certification of passive denitrification filters are requesting that, as part of the certification process, the provider also specify the aerobic treatment system(s) that would be acceptable to the provider as pretreatments for the denitrification filter (Loomis, 2007). If the treatment system under consideration already includes an aerobic treatment process, then addition of a passive denitrification filter could in itself provide substantially increased total nitrogen removal. The term recirculating extracted 26 citations from the Florida Passive Nitrogen Removal Study Citation List. Some references appears in more than one folder in the attached CD for the reason that they cover more than one subject classification or that they have subject common to more than one area. One example is citations in the Drainfield Modification folder. The organization framework of Table 2 is used in the following section to review the individual citations.

12

http://www.endnote.com/


Table 5 Organization of Citations in Electronic Files ( ) number of files Onsite Nitrogen Removal (10)

Aerobic Unsaturated Filters (Unsaturated)

Recirculating Sand Filters (6)

Peat Biofilters (9)

Open Cell Foam Biofilters (2)

Textile Biofilters (2)

Coir Biofilters (4)

Zeolite Biofilters (2)

Tire Chip (1)

Anoxic Filters (Saturated)

Heterotrophic Processes

Cellulosics (7)

Point (4)

Nitrex (5)

RI Systems (4)

La Pine Study (5)

Other Carbon Donors (6)

Autotrophic Processes

Sulfur (38)

Sulfide (1)

Iron (1)

Heterotrophic/Autotrophic Processes (3)

Drainfield Modifications (10)

Point (4)

Black & Gold (1)

Multi Soil Layers (5)

Soil Denitrification (1)

13


Review of Passive Nitrogen Removal Technologies with potential for application in passive on-site nitrogen removal systems are discussed here. Results of field and laboratory performance evaluations and experiments are summarized in Appendix C, Passive Nitrogen Technology. Nitrogen in STE occurs in reduced form as organic nitrogen or ammonia. Total nitrogen removal requires aerobic nitrification as a first biochemical reaction followed by denitrification. These must occur within process tanks, in natural systems, or within soil treatment units (drainfields) modified for enhanced nitrogen removal. The complete citation list for the literature review is contained in Appendix B and in the Endnote file that is an integral part of this report. In this section, tables are presented which contain the number designations for citations that refer to the citation list in Appendix B. Unit Operations As a biochemical necessity, ammonification and nitrification is required prior to passive denitrification filters. Removal efficiency and effluent concentrations of organic nitrogen and ammonia are of great concern in the initial aerobic stage, as well as the aerobic effluent water quality that could affect operation of the anoxic denitrification filter. Passive denitrification filters operate with lower dissolved oxygen or under completely anoxic conditions, and are limited in their ability to remove reduced nitrogen (i.e. organic and ammonia nitrogen). Initial treatment units that promote nitrification may also denitrify and reduce total nitrogen, and recirculation around the first stage (as in recirculating sand filters for example) can increase total nitrogen removal and lower the nitrate loading on subsequent passive denitrification filters. Recirculation around the aerobic treatment filter also restores alkalinity. In a single pass aerobic system, nitrification could result in a decline in pH due to alkalinity consumption. Inhibition of nitrification at lower pH could result in a deterioration in ammonia removal performance. The increasing emphasis on domestic water conservation could result in higher total nitrogen (TN) levels in septic tank effluent and increases in the TN/alkalinity ratio. The potential for inhibition of nitrification could be increased with water conservation. Factors that influence the selection of a passive nitrogen removal technology include the water quality characteristics of STE, target effluent nitrogen levels, and the desired treatment reliability. It should be realized that there may be limitations on the concept of a completely passive treatment system for removal of Total Nitrogen from onsite wastewater. For example, an inverse relationship may exist between nitrogen removal effectiveness and treatment system passivity. This relationship is not strongly defined. The literature review was conducted to examine currently employed and possibly new approaches to passive nitrogen removal, and to identify technologies and combinations of systems that could be used. Aerobic (Unsaturated) Filters Prominent nitrification processes include intermittent and recirculating sand filters, peat filters, textile filters, and filters with other media. These systems are summarized in Table 6. All systems contain porous media through which wastewater flows downward as a trickle flow over

14


15

media surfaces. Oxygen is supplied by ingress of air through pore spaces in the media. All systems are capable of substantial reductions of organic nitrogen and ammonia. A feature common to many unsaturated filters is enhancement of total nitrogen reduction by recirculation, which provides pre-denitrification using organic matter in the wastewater as the carbon source/electron donor. Summaries of unsaturated filter technologies have been presented in Jantrania and Gross (2006), Leverenz et al. (2002) and Crites and Tchobanoglous (1998). Recirculating sand filters (RSF) are capable of achieving ammonia removals of 98% and Total N removals of 40 to over 70% (Kaintz et al., 2004; Louden et al., 2004; Piluk and Peters, 1994; Richardson et al., 2004;). Effluent ammonia levels of 3 mg/L or less can be achieved (Urynowicz et al., 2007). Low temperatures have been suggested to adversely affect RSF ammonia removal performance, but adverse temperature effects should be of limited significance in the Florida climate. Peat filters can achieve ammonia nitrogen removal efficiencies of 96% or greater from septic tank effluent, with effluent NH3-N in some cases of 1 mg/L or less (Lacasse, 2001; Lindbo and MacConnell, 2001; Loomis et al., 2004; Patterson, 2004; Rich, 2007). TN reductions of 29 to 41% have been reported in modular recirculating peat filters (Monson Geerts et al., 2001a); 44% in peat filters using pressurized dosing (Patterson et al., 2004); and 15 and 21% in two single pass modular peat filters. Recirculating textile filters achieved 44 to 47% TN reduction (Loomis et al., 2004) from septic tank effluent. In some cases, textile filters treating septic tank effluent have produced effluents with NH3-N levels of less than 1 mg/L (Rich, 2007). Textile filters also produce nitrified effluents (McCarthy, et al., 2001; Rich, 2007; Wren et al. 2004) and are often operated at higher hydraulic loading rates (Table 6). The Waterloo Biofilter is a proprietary treatment system that has been demonstrated to reduce septic tank effluent TN by 62% while also providing over 90% ammonia N removal (132). Aerocell is another open cell foam media filter that operated with recycles and achieves 77% total nitrogen removal (Table 6). Tire crumb or tire chip has been employed as a substitute for gravel in disposal trenches, and has been summarized by Grimes et al (2003).


Table 6 Summary of Unsaturated Aerobic Media Filters

System Type Description Features Typical Performance Range Citations (Refer to Appendix B)

Intermittent sand filters

Sand filter Single pass

0.3 to 0.7 mm media 18 to 36 in. depth 0.7 to 1.5 gal/ft2-day 12 to 48 dose/day

TN Removal: 20 to 50% Effluent: 20 to 20 mg/L NH3-N Effluent: 1.9 to 9 mg/L

10,12,28,41,49, 58,64,71,88,94,111,134,169

Recirculating sand filters

Sand filter Recirculation

1.5 to 3 mm media 18 to 36 in. depth 3 to 5 gal/ft2-day 40 to 120 dose/day

TN Removal: 40 to 75% Effluent: 15 to 30 mg/L NH3-N Effluent: 1 to 5 mg/L

20,24,33,40,41,53, 56,84,88,89,94,111,118,130,131,142, 153,159,166,199, 210,209

Textile biofilters Textile filter Recirculation

2 to 3 in. cubes 36 to 72 in. depth 8 to 17 gal/ft2-day 80 to 140 dose/day

TN Removal: 20 to 60% Effluent: 10 to 60 mg/L NH3-N Effluent: 1.7 to 5.9 NO3-N Effluent: 11 mg/L

47,84,88,111,117, 123,158,218

Peat biofilters Peat media filter Single pass or recirculation

246 to 36 in. depth 3 to 6 gal/ft2-day 12 to 120 dose/day

TN Removal: 10 to 75% Effluent: 10 to 60 mg/L TKN Removal: 90 to 95% NH3-N Effluent: 1 mg/L NO3-N Effluent: 20 to 50

20,47,56,84,88,108,117,123,124,126, 127,147-149,158, 163,199,216

16


17

Table 6 Summary of Unsaturated Aerobic Media Filters (Continued)

System Type Description Features Treatment Performance Citations (Refer to Appendix B)

Waterloo biofilter Open cell foam media, single pass or recirculation

3 to 4 in. cube media 48 in. depth 11 gal/ft2-day

TN Removal: 62% Effluent: 14 mg/L NH3-N Effluent: 2.4 mg/L NO3-N Effluent: 10 mg/L

135

Zeolite biofilters Zeolite media filter 20 to 30 in. depth 6.1 gal/ft2-day

NH3-N Removal: 98.6% Influent: 70 mg/L Effluent: 1 mg/L NO3-N Effluent: 57 mg/L

151

Coir biofilters Coir filter bed, with recirculation

Coconut coir media 18 gal/ft2-day 5.88 gal/ft3-day

TN Removal: 55% Influent: 38 mg/L Effluent: 17 mg/L TKN Removal: 83% Influent: 38 mg/L Effluent: 6.5 mg/L

137,180,181,196

Aerocell biofilter Open cell foam media filter, with recirculation

2 in. cube media 18 gal/ft2-day 5.88 gal/ft3-day

TN Removal: 77 % Influent: 40 mg/L Effluent: 9.3 mg/L TKN Removal: 87% Influent: 40 mg/L Effluent: 5.4 mg/L

136


As a group, nitrification processes are reasonably well developed technologies. Synthetic media generally have lower footprints and higher areal hydraulic loading rates that traditional sand filters. Issues involved include the use and need for recirculation, effluent levels of organic and ammonia N achievable, and reliability of performance. With some media, recirculation may be more important to maintaining a given level of ammonia removal, which may be related to oxygen ingress into the site of action of attached nitrifying microorganisms. A rational basis for comparison of aerobic media could potentially be developed using the effective media surface area within the filter bed, as perhaps modified by factors effecting oxygen ingress and by recirculation. The overriding requirement for the aerobic treatment performance is to produce low effluent levels of organic N and ammonia N prior to treatment in anoxic reactive media filters. Factors affecting performance of unsaturated aerobic media filters are listed in Table 7. The hydraulic loading rate and loading rates of organics and nitrogen are important operating characteristics, particularly as they relate to the functioning of the physical and biological processes within the media. Key factors for successful treatment in an unsaturated media filter are surface area for attachment of microorganisms and for sorption of colloidal constituents in the wastewater, the need for sufficient pore space for assimilation of solids materials and their biodegradation between doses, the water retention capacity of the media, and the pore space that is available for aeration. The characteristics of media that influence performance of unsaturated filters are listed in Table 8. The performance of any unsaturated media filter is determined by the interactions of media characteristics (Table 8) with system parameters (Table 7). A significant interaction that occurs between the media and the system is the water retention capacity of media versus the hydraulic application rate. High water retention capacity is desirable to retain wastewater within the filter and achieve low effluent levels. The water retention capacity of media must exceed the hydraulic application rate per dose to prevent rapid movement of applied wastewater through the filter. More frequent doses (lower volume per dose), coupled with high water retention media, represent the most favorable combination. Another highly critical factor to optimum functioning of unsaturated media filters is the aeration pore space. Unsaturated media filters are four phase systems: solid media, attached microbial film, percolating wastewater, and gas phase. The total porosity (excluding internal pore spaces within the media) must be shared between attached biofilm, percolating water, and gas phase. A media with a high total porosity will more likely allow sufficient oxygen transfer throughout the filter bed, providing more effective utilization of the total media surface area for aerobic treatment. If media size becomes too small, a larger fraction of the pores may remain saturated and become inaccessible to oxygen transfer. For example, sand with a total porosity of 38% could have an aeration porosity of only 2.5% of the total media volume, depending on sand size and the hydraulic application rate. Such conditions could decrease nitrification effectiveness and perhaps also increase denitrification within microzones with limited contact with the

18


Table 7 Factors Influencing Performance of Unsaturated Aerobic Filters

Feature Effect

Hydraulic loading rate Higher rates lower water retention time and treatment

Organic loading rate Higher loading rates increase rate at which biofilms must process organic matter; nitrification may be inhibited of too high

Nitrogen loading rate Higher loading rates require higher nitrification rates and higher oxygen utilization rates

Media depth Deeper beds can give better treatment; uppers layers often more reactive

Specific surface area Higher values give greater attachment surfaces for microorganisms

Superficial velocity Effects mass transfer between wastewater and biofilms

Average linear velocity Effects mass transfer between wastewater and biofilms

Hydraulic application rate per dose Volume per dose should be scaled to field capacity of media Organic loading rate per dose Loading per dose must not exceed processing rate Nitrogen loading rate per dose Loading per dose must not exceed processing rate

Average water residence time Longer residence time gives more time for biochemical reactions and better treatment

Uniformity of Dosing Promotes full utilization of all elements of the filter media

Wastewater

Suspended solids Accumulated within pores, may lead to clogging if not biodegraded

BOD High values require more room for attached growth and metabolism between doses, particularly in upper filter layers

Organic and ammonia nitrogen Significant component of total oxygen supply requirement

Alkalinity Consumed by nitrification and restored by heterotrophic denitrification; adequate supply needed to prevent pH decline by nitrification

19


Table 8 Media Characteristics Influencing Performance of Filters

Feature Effect

Particle size distribution Larger particles less subject to clogging Smaller particles have greater surface area per volume for treatment

Uniformity coefficient Effects flow uniformity

Specific surface area Higher values give greater attachment surfaces for microorganisms

Air filled porosity Oxygen supply throughout media depth for BOD oxidation and nitrification in unsaturated filters

Water retention capacity

Higher water retention in unsaturated media filters provides longer time of contact of water with microorganisms and better treatment; affected by intrinsic porosity that favors capillary water retention

Sinuosity and tortuosity Affect accessibility of pore spaces to exchange of wastewater and air

Specific weight Effects compression strength required for support in multi media filters Ion exchange capacity Ammonia adsorption may improve performance

Compressibility Effects material resistance to compression when wetted with biofilm and attached solids

Biodegradation Biodegradation of organic media will limit longevity

Resilience Prevents compaction under deployment

Hydrophilicity Attracts water for wetting and rewetting

20


gas phase. Denitrification within an unsaturated filter would improve total nitrogen removal but could result in less efficient nitrification and higher effluent ammonia concentrations. By contrast, media with high total porosity would be more likely to have a sufficiently high aeration porosity to allow effective utilization of all media surface area and better ammonia removal performance. If the goal is to achieve total nitrogen removal in an overall system containing an unsaturated filter followed by an anoxic, reactive media denitrification filter, then the goal of low effluent ammonia should take precedence over denitrification in the unsaturated first stage filter. An example media with high total porosity and high water retention capability is sphagnum peat moss. The total porosity of sphagnum peat is greater than 85%, and percolating water might occupy two thirds of this available pores. Under these conditions, pore space available for aeration would be over 25% of the total volume of the filter bed. The very low effluent ammonia levels that peat filters appear capable of producing may be related to these factors. Media with significant ion exchange capacity may offer a method to superior removal of ammonia nitrogen in flowing systems (Philip and Vasel, 2006; Smith, 2006). Zeolite media are excellent surface for biofilm attachment, and have relatively high porosities. Sorption of ammonium ions onto zeolite media can sequester ammonium ions from the water and provide enhanced contact with attached nitrifying organisms under steady flow conditions. Sorption also provides a buffer when loading rates are high or other factors inhibit nitrifier activity, resulting in increased resiliency of the treatment process. Ammonia ion exchange adsorption onto zeolites is reversible, and microorganisms can biologically regenerate the zeolite media in periods of lower loading. A zeolite filter for onsite wastewater treatment removed 98.6% of ammonia and produced an effluent ammonia nitrogen concentration of 1 mg/L when operated at 6.1 gal/ft2-day (Philip and Vasel, 2006). Other bench scale and pilot studies have demonstrated the ability of zeolite filters to maintain high ammonia removal under high non-steady loadings of ammonia nitrogen (Smith, 2006). Coconut coir is a natural, renewable material that is a waste product from coconut production. Coir has many of the same properties of peat that make it a desirable treatment media, including high surface area, high water retention, and high porosity (Talbot, 2006), and has been successfully used as a planting media in greenhouses. While most coir is produced in Asia, Florida contains abundant coconut palm trees that could potentially provide a sustainable material source. A onsite wastewater treatment system using coconut coir has been reported (Sherman, 2006: Sherman, 2007). Synthetic fiber materials could have many of the same advantages as a media as coir. Candidate media for the unsaturated media filter should possess many of the desirable characteristics that have been discussed above. Zeolite filters also have promise for unsaturated flow filters for passive systems. The interaction of cation exchange media with microbial reactions appears to offer potential for passive treatment with enhanced performance. Other candidate media include expanded clays, expanded shales, and tire crumb.

21


Anoxic (Saturated) Filters Anoxic saturated media filters form a second stage in the passive nitrogen removal system. The anoxic filters contain a “reactive” media that provide a slowly dissolving source of electron donor for reduction of nitrate and nitrite by microbial denitrification. Denitrifying microorganisms grow predominantly attached to the media surfaces. Water flows by advection through the media pores, where the oxidized nitrogen species is consumed by attached microorganisms. Water saturation of the pores prevents ingress of oxygen, which could interfere with nitrate reduction. Factors influencing the performance of anoxic denitrification filters are listed in Table 9. Hydraulic and nitrogen loading rates, surface area of media, pore size, and flow characteristics within the reactor are important considerations. The media is consumed by dissolution, and this process must be sufficiently rapid to supply electron equivalents for nitrate reduction and other possible reactions. On the other hand, rapid dissolution would reduce the longevity of the media. Too rapid a dissolution rate could also lead to the presence of excess dissolution products in the effluent (BOD for wood-based filters; sulfate for sulfur based filters). An aerobic process effluent low in BOD and suspended solids would be less likely to lead to channeling within the anoxic filter. Geometry of the column could affect flow patterns and potential channeling; the later effects could be overcome by use of larger systems. The effects of flow channeling on performance deterioration could require maintenance or media replacement at time scales appreciably shorter than longevities based on theoretical stoichiometric requirements of electron donor for denitrification. A summary of performance of passive anoxic denitrification filters is shown in Table 10. Heterotrophic Denitrification Passive heterotrophic denitrification systems use solid phase carbon sources including woodchips (Cooke et al., 2001; Greenan et al., 2006; Jaynes et al., 2002; Kim et al., 2003; Robertson et al., 2000; Robertson and Cherry, 1995; Robertson et al., 2005; van Driel et al., 2006), sawdust (Eljamal et al., 2007; Greenan et al., 2006; Jin et al., 2006; Kim et al., 2003; van Driel et al., 2006), cardboard (Greenan et al., 2006), paper (Jin et al., 2006; Kim et al., 2003), and agricultural residues (Cooke et al., 2001; Greenan et al., 2006; Jin et al., 2006; Kim et al., 2003; Ovez, 2006; a, 2006b). Limited studies have also been conducted using other carbon sources such as cotton (Della Roca et al., 2005), poly(e-caprolactone) (Horiba et al., 2005), and bacterial polyesters (Mergaert et al., 2001). Cellulosic-based systems using wood are the most developed heterotrophic denitrification filter technology. The Nitrex process uses a proprietary media containing woodchips and other materials (EPA, 2007; NSF, 2003; Lombardo, 2005; Robertson et al., 2000; Robertson and Cherry, 1995; Robertson et al., 2005). Several Nitrex demonstration studies have been conducted, which have followed sand or peat filters, and some have operated for greater than two years (Lombardo, 2005). Combined RSF/Nitrex systems have

22


23

Table 9 Factors Influencing Performance of Saturated Anoxic Filters

Feature Effect

Hydraulic loading rate Higher rates lower water retention time and treatment

Organic loading rate Higher loading rates increase rate at which heterotrophic biomass could accumulate

Solids loading rate Higher loading rates increase rate at which solids could accumulate

Nitrogen loading rate Higher loading rates require higher denitrification rates and higher rates of electron donor dissolution

Media depth Deeper beds can give better treatment; uppers layers often more reactive

Specific surface area Higher values give greater surface area for attachment of microorganisms and dissolution of media

Superficial velocity Effects mass transfer between wastewater and biofilms

Average linear velocity Effects mass transfer between wastewater and biofilms

Average water residence time Longer residence time gives more time for biochemical reactions and better treatment

Wastewater

Suspended solids Accumulated within pores, may lead to preferential flow if not biodegraded

BOD Will create more heterotrophic biomass and may increase potential for preferential flow

Nitrate nitrogen High loadings require greater surface areas and higher levels of denitrifying activity

Alkalinity Consumed by autotrophic denitrification; must be balanced by sum of influent alkalinity and alkalinity provided by solid source


Table 10 Summary of Saturated Anoxic Media Filters


Sulfur/oyster shell filter (bench scale)

1 liter bench column synthetic wastewater upflow single pass

Sulphur/oyster shell media (75/25% by volume) Sulphur: 4.7 mm

anoxic only

NO3-N Removal: 80% Influent: 50 mg/L Effluent: 10 mg/L

173

Sulfur/oyster shell filter

185 gal. column aerobic effluent upflow single pass

Sulphur/oyster shell media (75/25% by volume) 47 gal/ft2-day

anoxic only

TN Removal: 82% Effluent: 4.2 mg/L NO3-N Removal: 88% Influent: 20 mg/L Effluent: 2.4 mg/L

23

Sulfur/limestone column

237 gal. column groundwater upflow single pass Residence time: 13 hr.

Sulphur/limestone media (67/33% by volume) 63 gal/ft2-day Sulfur: 2.5 to 3.0 mm Limestone: 2.38 to 4.76 mm

anoxic only

NO3-N Removal: 96% Influent: 64 mg/L

Effluent: 2.4 mg/L NO2-N Effluent: 0.2 mg/L

46

24


25

Table 10 Summary of Saturated Anoxic Media Filters (Continued)


NitrexTM

aerobic effluent gravity flow upflow single pass

Nitrex wood-based media 24 to 30 inch media depth (est.) 4.6 gal/ft2-day (est.)

aerobic+anoxic

TN Removal: 79 to 96% Effluent: 3 to 18 mg/L NO3-N Effluent: 0.3 to 8 mg/L

54,62,114,116, 158,160,162,203

Black& GoldTM

wood-based media single pass downflow gravity

Influent: STE 280 gal. column Sand/tire crumb/woodchip (85/11/5% by volume) 8.3 gal/ft2-day

aerobic+anoxic

TN Removal: 98% Influent: 414 mg/L Effluent: 7.1 mg/L NH3-N Effluent: 4.4 mg/L NO3-N Effluent: 0.05 mg/L

176


produced average TN removals of 88 to 99% from septic tank effluent, with average effluent NO3-N concentrations of 1 to 2 mg/L. In another study, a subsurface leaching chamber was installed beneath an active parking lot for on-site sewage treatment, using sawdust as carbon source (St. Marseille and Anderson, 2002). At a loading of 1.22 gallons/ft2-day; the effluent NO3-N averaged 0.6 mg/L. Other heterotrophic denitrification systems have been successfully tested at laboratory scale. Factors that affect the long term success of carbon-based denitrification filters include the long term availability of carbon supply for the wastestream being treated and the physical structure of the biodegradable components of the media. As for any packed bed, biologically active media filter which is deployed over extended periods of time, the long term hydraulics of the unit are a possible issue. Accumulation of biological and inorganic solids could lead over time to the development of preferential flow paths within the filter, reducing average residence time and wastewater contact with the media. The result of flow short circuiting would be performance deterioration. The practical aspects of media replacement and management/disposal must be considered, in light of the frequency with which media must be supplemented or replaced. Another factor is the release of soluble biodegradable carbon as water passes through the filter, which could increase biochemical oxygen demand (BOD) and chemical oxygen demand (COD). It is possible that this material would be readily consumed within tens of feet of release in a groundwater plume, or within a solid treatment unit receiving the effluent of the carbon-based denitrification filter. Autotrophic Denitrification The autotrophic denitrification systems that have received the most attention are elemental sulfur-based media filters, which are under development. Sulfur-based denitrification filters have employed limestone or oyster shell as a solid phase alkalinity source to buffer the alkalinity consumption of the sulfur-based biochemical denitrification (Brighton, 2007; Darbi et al., 2003a, 2003b; Flere and Zhang, 1998; Kim et al., 2003; Koenig and Liu, 2002; Nugroho et al., 2002; Sengupta and Ergas, 2006; Sengupta et al. 2007; Sengupta et al., 2006; Shan and Zhang, 1998; Zeng and Zhang, 2005; Zhang, 2002; Zhang, 2004). A pilot scale filter containing elemental sulfur and oyster shell at a 3:1 ratio was operated for 11 months at the Massachusetts Alternative Septic System Test Center (Brighton, 2007). The filter received the effluent from a Clean Solution aerobic treatment system that was receiving septic tank effluent. TN was reduced 82% through the sulfur/oyster shell filter, while the aerobic/sulfur filter system removed 89.5% TN from the septic tank effluent. A pilot scale elemental sulfur/limestone column was operated for 6 months on a well water containing 65 mg/L NO3-N; nitrate removal averaged 96% and average effluent NO3-N was 2.4 mg/L (Darbi et al., 2003a). A laboratory sulfur/oyster shell column was operated at an Empty Bed Contact Time of 0.33 to 0.67 days and removed 80% of influent nitrate (Sengupta and Ergas, 2006; Sengupta et al., 2006). Some factors that affect the long term performance success of autotrophic denitrification filters are similar to those for carbon-based denitrification filters. They include the long term availability of electron donor supply for the wastestream being treated, and the physical structure of the biodegradable components of the media. Versus wood based organics electron donors, elemental sulfur could possibly remain physically intact for longer time periods. As for any

26


packed bed, biologically active media filter deployed over extended periods of time, the long term hydraulics of the unit are a concern. Accumulation of biological and inorganic solids could lead over time to the development of preferential flow paths within the filter, reducing average residence time and wastewater contact with the media. To the extent that these processes occur, deterioration of performance could result. The timescales of media replacement, maintenance and supplementation and the practical aspects of these activities must be considered. Another factor is the release of sulfate as water passes through the filter, and possible odors through hydrogen sulfide generation. Several candidate media can be suggested for the saturated media filter which forms the second stage of a passive onsite nitrogen removal system for Florida. Media should possess many of the desirable characteristics that have been previously discussed. Both elemental sulfur and wood based treatment systems are readily available and economical candidates. Crushed oyster shell is readily available. These alkalinity sources could also be used in a single pass, unsaturated first stage filter if nitrification would otherwise be inhibited. The interaction of cation exchange media with microbial reactions appears to offer potential for passive treatment with enhanced performance. Expanded shales with anion exchange capacity are commercially available and could be used in mixed media to increase the resiliency and performance of second stage anoxic denitrification filters. Drainfield Modifications Modifications to drainfields entail the in-situ addition of a permeable media that supports denitrification through the release of carbon or electron donor. Wastewater (septic tank effluent) would initially pass through an unsaturated layer or zone (of sand for example), where nitrification occurs. Following passage through the unsaturated zone, the wastewater would pass through a permeable denitrification layer or zone. Denitrification media could be placed as an underlayment beneath the unsaturated soil, or as a subdivided treatment zone within a drainfield through which effluent from the aerobic zone must pass. A modified drainfield design using a sulfur/limestone layer beneath a sand layer provided greater than 95% TN removal in laboratory scale columns receiving primary effluent from a municipal wastewater treatment plant (Shan, 1998). Nitrification occurred in the upper sand layer, and the lower denitrification layer was not maintained in a saturated condition.

florida passive nitrogen removal study · present worth (pw) and uniform annual cost (uac) were...

Documents